Should be back 24/7 soon!

[Stonesthroen]

It has been a really hot summer but it is starting to get cold again so i should be able to start crunching 24/7 with my main rig again pretty soon. Hopefully can pass the 1 million mark in the next week as i have been stuck to one machine for the last few months due to the heat!!

 

[baditude_df]

Excellent to hear. Thank you!

 

[muddocktor]

And I just shut down my heatsink test system from Rosetta. Gotta get it ready for some more heatsink testing and I don't want to have it full of work during testing. It took around 4 days to burn through all the work on it

 

[neonblingbling]

I'm starting back up as well, but for some reason I'm only getting a 'communication deferred' message when trying to get new work. Typical for being down for a while?

EDIT: Nevermind, it is going now!

 

[muddocktor]

Watch the work units you get from Rosetta for a while folks. I got some on my main system that were thrashing my hard drive to death and slowing everything down to a crawl on my computer. The hard drive indicator light was staying lit continuously while the Rosetta work was processing. If I snoozed the client or I closed out BOINC, the thrashing would eventually stop (when the selection could work it's way through the thrashing. I ended up having to dump all the work units on that machine to get it to quit thrashing my hard drive. I then set it back to get more work, but the Rosetta server would only send it 8 wus, which it gobbled up pretty quickly and it won't send me any more for another 15 hours. So my main rig is now doing some Seti work. If I get more trash wu's that attack my hardware like that, I'm out of this project for a while. I surely don't mind letting my systems do DC work for a project, but when they send out crap that ties my systems in knots, then they don't deserve my help.

BTW, the Seti work that system is doing isn't thrashing the hard drive like those defective Rosetta wus.

 

[Bobnova]

The rosetta WUs seem to be getting steadily larger, most of 'em seem to be in the 400mb range now, so if you're already using a couple GB on windows and firefox and then have eight rosetta threads at 400mb each, it's page file time!

 

[baditude_df]

I was fine for a few months, and now I'm getting a rash of stalled WU's. When I get up in the morning or get back from work, I'll find that my processor usage is at 75% or 50% or 25% or 0% and I'll have to go in to BOINC manager and abort the stalled WU's typically sitting at 2.5hrs or more run time. I've yet to figure out what the heck is going on with this machine and stalling WU's but it's been happening ever since I built it.

 

[PenguinsPal]

I've had to shut rosetta down on 1 of my PCs because it brings everything to s screeching halt.

 

[muddocktor]

PenguinsPal, I've been doing some experimentation. Go into preferences on the client on that machine and set the swap file usage to 0%. That stopped the hard drive thrashing for my main rig.

 

[PenguinsPal]

I'll give it a shot. It was heartbreaking to have to shut it down completely. Especially since I am so close to my 1,000,000 mark.